GTA_Assignments

This repository contains two simple Python implementations for:

• PageRank algorithm
• Node2Vec algorithm

PageRank Implementation in Python

It demonstrates how PageRank can be computed iteratively using a transition matrix and the power iteration method.

📌 Features

• Pure Python + NumPy implementation of PageRank
• Adjustable damping factor
• Iterative computation with convergence criterion
• Example with adjacency matrix input
• Prints the PageRank values and number of iterations to convergence

Node2Vec from Scratch with Skip-gram

Minimal Python implementation of node2vec embeddings for small graphs, using random walks and a Skip-gram model with Negative Sampling (SGNS).

---

📌 Features

• Build random walks on arbitrary graphs with biased sampling (p, q parameters).
• Extract center-context pairs from walks.
• Train Skip-gram with Negative Sampling using PyTorch.
• Obtain low-dimensional embeddings for nodes.
• Visualize node embeddings

---

🛠 Requirements

The implementations relies on the following Python packages:

• numpy
• pandas
• matplotlib
• torch
• networkx
• random

You can install them with:

pip install -r requirements.txt